Add mp4 support

libmat2/video.py

@@ -2,10 +2,37 @@ import os
 import subprocess
 import logging
 
+from typing import Dict, Union
+
 from . import exiftool
 
 
-class AVIParser(exiftool.ExiftoolParser):
+class AbstractFFmpegParser(exiftool.ExiftoolParser):
+    """ Abstract parser for all FFmpeg-based ones, mainly for video. """
+    def remove_all(self) -> bool:
+        cmd = [_get_ffmpeg_path(),
+               '-i', self.filename,      # input file
+               '-y',                     # overwrite existing output file
+               '-map', '0',              # copy everything all streams from input to output
+               '-codec', 'copy',         # don't decode anything, just copy (speed!)
+               '-loglevel', 'panic',     # Don't show log
+               '-hide_banner',           # hide the banner
+               '-map_metadata', '-1',    # remove supperficial metadata
+               '-map_chapters', '-1',    # remove chapters
+               '-disposition', '0',      # Remove dispositions (check ffmpeg's manpage)
+               '-fflags', '+bitexact',   # don't add any metadata
+               '-flags:v', '+bitexact',  # don't add any metadata
+               '-flags:a', '+bitexact',  # don't add any metadata
+               self.output_filename]
+        try:
+            subprocess.check_call(cmd)
+        except subprocess.CalledProcessError as e:
+            logging.error("Something went wrong during the processing of %s: %s", self.filename, e)
+            return False
+        return True
+
+
+class AVIParser(AbstractFFmpegParser):
     mimetypes = {'video/x-msvideo', }
     meta_whitelist = {'SourceFile', 'ExifToolVersion', 'FileName', 'Directory',
                       'FileSize', 'FileModifyDate', 'FileAccessDate',
@@ -24,25 +51,55 @@ class AVIParser(exiftool.ExiftoolParser):
                       'SampleRate', 'AvgBytesPerSec', 'BitsPerSample',
                       'Duration', 'ImageSize', 'Megapixels'}
 
+class MP4Parser(AbstractFFmpegParser):
+    mimetypes = {'video/mp4', }
+    meta_whitelist = {'AudioFormat', 'AvgBitrate', 'Balance', 'TrackDuration',
+                      'XResolution', 'YResolution', 'ExifToolVersion',
+                      'FileAccessDate', 'FileInodeChangeDate', 'FileModifyDate',
+                      'FileName', 'FilePermissions', 'MIMEType', 'FileType',
+                      'FileTypeExtension', 'Directory', 'ImageWidth',
+                      'ImageSize', 'ImageHeight', 'FileSize', 'SourceFile',
+                      'BitDepth', 'Duration', 'AudioChannels',
+                      'AudioBitsPerSample', 'AudioSampleRate', 'Megapixels',
+                      'MovieDataSize', 'VideoFrameRate', 'MediaTimeScale',
+                      'SourceImageHeight', 'SourceImageWidth',
+                      'MatrixStructure', 'MediaDuration'}
+    meta_key_value_whitelist = {  # some metadata are mandatory :/
+        'CreateDate': '0000:00:00 00:00:00',
+        'CurrentTime': '0 s',
+        'MediaCreateDate': '0000:00:00 00:00:00',
+        'MediaLanguageCode': 'und',
+        'MediaModifyDate': '0000:00:00 00:00:00',
+        'ModifyDate': '0000:00:00 00:00:00',
+        'OpColor': '0 0 0',
+        'PosterTime': '0 s',
+        'PreferredRate': '1',
+        'PreferredVolume': '100.00%',
+        'PreviewDuration': '0 s',
+        'PreviewTime': '0 s',
+        'SelectionDuration': '0 s',
+        'SelectionTime': '0 s',
+        'TrackCreateDate': '0000:00:00 00:00:00',
+        'TrackModifyDate': '0000:00:00 00:00:00',
+        'TrackVolume': '0.00%',
+    }
+
     def remove_all(self) -> bool:
-        cmd = [_get_ffmpeg_path(),
+        logging.warning('The format of "%s" (video/mp4) has some mandatory '
-               '-i', self.filename,      # input file
+                        'metadata fields; mat2 filled them with standard data.',
-               '-y',                     # overwrite existing output file
+                        self.filename)
-               '-loglevel', 'panic',     # Don't show log
+        return super().remove_all()
-               '-hide_banner',           # hide the banner
+
-               '-codec', 'copy',         # don't decode anything, just copy (speed!)
+    def get_meta(self) -> Dict[str, Union[str, dict]]:
-               '-map_metadata', '-1',    # remove supperficial metadata
+        meta = super().get_meta()
-               '-map_chapters', '-1',    # remove chapters
+
-               '-fflags', '+bitexact',   # don't add any metadata
+        ret = dict()  # type: Dict[str, Union[str, dict]]
-               '-flags:v', '+bitexact',  # don't add any metadata
+        for key, value in meta.items():
-               '-flags:a', '+bitexact',  # don't add any metadata
+            if key in self.meta_key_value_whitelist.keys():
-               self.output_filename]
+                if value == self.meta_key_value_whitelist[key]:
-        try:
+                    continue
-            subprocess.check_call(cmd)
+            ret[key] = value
-        except subprocess.CalledProcessError as e:
+        return ret
-            logging.error("Something went wrong during the processing of %s: %s", self.filename, e)
-            return False
-        return True
 
 
 def _get_ffmpeg_path() -> str:  # pragma: no cover

mat2

@@ -20,6 +20,8 @@ __version__ = '0.5.0'
 assert Tuple
 assert Union
 
+logging.basicConfig(format='%(levelname)s: %(message)s', level=logging.WARNING)
+
 
 def __check_file(filename: str, mode: int=os.R_OK) -> bool:
     if not os.path.exists(filename):

tests/test_libmat2.py

@@ -521,3 +521,26 @@ class TestCleaning(unittest.TestCase):
         os.remove('./tests/data/dirty.cleaned.zip')
         os.remove('./tests/data/dirty.cleaned.cleaned.zip')
 
+
+    def test_mp4(self):
+        try:
+            video._get_ffmpeg_path()
+        except RuntimeError:
+            raise unittest.SkipTest
+
+        shutil.copy('./tests/data/dirty.mp4', './tests/data/clean.mp4')
+        p = video.MP4Parser('./tests/data/clean.mp4')
+
+        meta = p.get_meta()
+        self.assertEqual(meta['Encoder'], 'HandBrake 0.9.4 2009112300')
+
+        ret = p.remove_all()
+        self.assertTrue(ret)
+
+        p = video.MP4Parser('./tests/data/clean.cleaned.mp4')
+        self.assertNotIn('Encoder', p.get_meta())
+        self.assertTrue(p.remove_all())
+
+        os.remove('./tests/data/clean.mp4')
+        os.remove('./tests/data/clean.cleaned.mp4')
+        os.remove('./tests/data/clean.cleaned.cleaned.mp4')